Find best Ready-made Analyzer/Tokenizer/TokenFilters for Solr that divide input string

Question

I am moving some existing Index from Lucene to Solr. We apply the following Logic on the Input text in Lucene:

1. to lower case
2. replaceDictionaryWords (replace some specific words by other words example replace "hertz" by "htz")
3. extract characters and digits only
4. trim output string
5. replace \s+ by \s
6. split using java.lang.String#split(in) method
7. for each splitted text, divide the result word by the following pattern: "ABCDEF" => ABC BCD CDE DEF (divide on 3, 2)

I don't want to write Tokenizer that might be exist.

So, I looked here http://wiki.apache.org/solr/AnalyzersTokenizersTokenFilters but get missed.

Answer 1

1. LowerCaseFilter,
2. SynonymFilter,
3. StandardTokenizer or PatternTokenizer,
4. TrimFilter,
5. PatternReplaceFilter,
6. WordDelimiterFilter?
7. NGramTokenFilter (you may need to write a factory for this one).

But if you already have an existing Lucene analyzer, you can make Solr use it.

Answer 2

Try OpenPipeline. It's designed for preprocessing documents that get fed to search software.